The overarching goal of this R21 proposal is to screen for lead compounds to target metabolic vulnerability of PIK3CA-mutant colorectal cancer. PIK3CA, which encodes the p110? catalytic subunit of PI3 kinase, is frequently mutated in a variety of human cancers including 20 to 30% of colorectal cancers. Despite huge effort made by the pharmaceutical industry and academic institutions in the last decade, a specific therapeutic approach targeting PIK3CA mutant cancers remains to be developed. We recently discovered that PIK3CA mutations render CRC addicted to glutamine through up-regulation of glutamate pyruvate transaminase 2. We demonstrate that inhibition of GPT2 by either shRNA or a tool compound selectively suppresses in vivo xenograft tumor PIK3CA-mutant, but not PIK3CA WT colorectal cancers. Our studies demonstrated that GPT2 is a critical target for PIK3CA mutant colorectal cancers. Here we propose to perform a high throughput screen (HTS) to identify GPT2-specific inhibitor. We have successfully developed a robust high throughput GPT2 enzymatic assay (z' 0.7-0.8). In aim 1, we will perform a HTS on the NCATS Genesis chemical library (~100,000 drug-like compounds). In aim 2, we will further characterize hits identified from the HTS to develop GPT2-specific inhibitors through the following sequential steps: (1) assay GPT2 binding affinity; (2) determine on-target effect of the hits in colorectal cancer cells; (3) determine selectivity of the hits to GPT2; (4) initially optimize the lead compounds using structure-activity relationship (SAR)-by-catalog; and (5) assess tumor inhibitory effect of the optimized GPT2 inhibitors in PIK3CA mutant colorectal cancer cells. Colorectal cancer is the second leading cause of cancer death and remains an unmet need for therapeutic development. We propose here to screen for lead compounds to be developed as potent and specific GPT2 inhibitors for treating PIK3CA mutant colorectal cancers. Given that PIK3CA is mutated in 20 to 30% of colorectal cancers, successful completion of our studies will have a huge impact on colorectal cancer therapy. GPT2 is a novel target uncovered by us and GPT2-specific inhibitors have never been developed before. Our proposed study will provide the first GPT2-specific inhibitors. Lastly, given that PIK3CA is the most frequently mutated oncogene in human cancers, the impact of our studies can extend beyond colorectal cancer.